scholarly journals Uncovering Complex Relations in Patient Pathways based on Statistics: the Impact of Clinical Actions

2022 ◽  
Author(s):  
Jelmer Jan Koorn ◽  
Xixi Lu ◽  
Felix Mannhardt ◽  
Henrik Leopold ◽  
Hajo Alexander Reijers
Keyword(s):  
2021 ◽  
Author(s):  
Marco Luca Sbodio ◽  
Natasha Mulligan ◽  
Stefanie Speichert ◽  
Vanessa Lopez ◽  
Joao Bettencourt-Silva

There is a growing trend in building deep learning patient representations from health records to obtain a comprehensive view of a patient’s data for machine learning tasks. This paper proposes a reproducible approach to generate patient pathways from health records and to transform them into a machine-processable image-like structure useful for deep learning tasks. Based on this approach, we generated over a million pathways from FAIR synthetic health records and used them to train a convolutional neural network. Our initial experiments show the accuracy of the CNN on a prediction task is comparable or better than other autoencoders trained on the same data, while requiring significantly less computational resources for training. We also assess the impact of the size of the training dataset on autoencoders performances. The source code for generating pathways from health records is provided as open source.


2012 ◽  
Vol 17 (1_suppl) ◽  
pp. 23-30 ◽  
Author(s):  
Pauline Allen ◽  
Simon Turner ◽  
Will Bartlett ◽  
Virginie Perotin ◽  
Greenwell Matchaya ◽  
...  

Objectives To assess the impact of provider diversity on quality and innovation in the English NHS by mapping the extent of diverse provider activity and identifying the differences in performance between Third Sector Organisations (TSOs), for-profit private enterprises, and incumbent organizations within the NHS, and the factors that affect the entry and growth of new providers. Methods Case studies of four local health economies. Data included: semi-structured interviews with 48 managerial and clinical staff from NHS organizations and providers from the private and third sector; some documentary evidence; a focus group with service users; and routine data from the Care Quality Commission and Companies House. Data collection was mainly between November 2008 and November 2009. Results Involvement of diverse providers in the NHS is limited. Commissioners' local strategies influence degrees of diversity. Barriers to entry for TSOs include lack of economies of scale in the bidding process. Private providers have greater concern to improve patient pathways and patient experience, whereas TSOs deliver quality improvements by using a more holistic approach and a greater degree of community involvement. Entry of new providers drives NHS trusts to respond by making improvements. Information sharing diminishes as competition intensifies. Conclusions There is scope to increase the participation of diverse providers in the NHS but care must be taken not to damage public accountability, overall productivity, equity and NHS providers (especially acute hospitals, which are likely to remain in the NHS) in the process.


Blood ◽  
2009 ◽  
Vol 114 (22) ◽  
pp. 352-352 ◽  
Author(s):  
Gareth J Morgan ◽  
Faith E Davies ◽  
Walter M Gregory ◽  
Susan E Bell ◽  
Alex J Szubert ◽  
...  

Abstract Abstract 352 We present updated results from MRC Myeloma IX study evaluating the role of the addition of thalidomide to the induction and maintenance of patients with myeloma. The study ran from May 2003 – November 2007 and randomised 1,970 patients and now has a median follow up of more than 3.5 years giving it improved power to detect changes in outcome developing later after treatment. Projected median OS younger fitter patients 66 months, median OS older less fit patients 32 months. The trial comprised of 2 patient pathways, one for younger fitter patients comparing CTD (cyclophosphamide, thalidomide, dexamethasone) with CVAD (cyclophosphamide, vincristine, adriamycin, dexamethasone), all patients going on to receive an ASCT – median age 59 years. In older less fit patients, melphalan and prednisolone (MP) was compared to CTD attenuated – median age 73 years. In both pathways following initial treatment, eligible patients were randomised to low-dose thalidomide or no maintenance. Patient's response was monitored using electrophoresis, serum free light chain and multiparameter flow cytometry. Cytogenetics was availabel on up to 60% of cases and gene expression on a subset of these. CTD is a well tolerated regimen with a good safety profile giving excellent survivals in both groups of patients despite a small increase in risk of VTE. Using modified EBMT criteria, the addition of thalidomide to induction treatment increases both response rates and depth of response for all age groups. Preliminary results as follows: overall response: CTD vs CVAD: 91% v 82%; CR 21% v 14% and 100 days post-HDM, better responses were seen in CTD with CR rates 65% v 48%. Remission depth was also greater in CTD with more patients achieving minimal residual disease negativity by flow cytometry. The addition of thalidomide increases response rates overall, and particularly complete response (CR) rates (a 17% increase in CR rates post HDM, p=.006). In older/less fit patients CTDa vs MP: overall response 83% v 46%; CR 21% v 4%. Definitive results of these analyses will be presented as well as how they translate into PFS and OS and by cytogenetic subgroup. There is a substantial increase in response with the inclusion of thalidomide but at a median follow-up of three years we are not as yet seeing a substantial increase in survival in either of the two broad patient groups. We have collected data on treatment at relapse to explore how this confounds OS data. Importantly modelling analyses indicate when and to what extent, with further follow-up, the survival differences that should accrue from this increase in CR rate are likely to translate into a survival benefit. These results have a number of important implications. We show the benefit of the addition of thalidomide to myeloma treatment but also highlight the importance of later analysis of such trials because of the emergence of significant changes at these later time points. We will present full updated results from the study including the impact of thalidomide on cytogenetic subgroups and in the maintenance setting. Disclosures: No relevant conflicts of interest to declare.


2021 ◽  
Author(s):  
Arabella Scantlebury ◽  
Catriona McDaid ◽  
Stephen Brealey ◽  
Elizabeth Cook ◽  
Arun Ranganathan ◽  
...  

Abstract Background: Recruitment of patients is one of the main challenges when designing and conducting randomised controlled trials (RCTs). Trials of rare injuries, or those that include surgical interventions pose added challenges due to the small number of potentially eligible patients and issues with patient preferences and surgeon equipoise. We explore key issues to consider when recruiting to orthopaedic surgical trials from the perspective of staff and patients with the aim of informing the development of strategies to improve recruitment in future research. Design: Two qualitative process evaluations of a United Kingdom-wide orthopaedic surgical RCT (ACTIVE) and mixed methods randomised feasibility study (PRESTO). Qualitative semi-structured interviews were conducted and data was analysed thematically. Setting: NHS secondary care organisations throughout the United Kingdom. Interviews were undertaken via telephone. Participants: 37 trial recruiters (surgeons, physiotherapists, research nurses) and 22 patients were interviewed. Patients included those who agreed to participate in the ACTIVE and PRESTO studies and patients that declined participation in the ACTIVE study. Results: We used a mixed methods systematic review of recruiting patients to randomised controlled trials as a framework for reporting and analysing our findings. Our findings mapped onto those identified in the systematic review and highlighted the importance of equipoise, randomisation, communication, patient’s circumstances, altruism and trust in clinical and research teams. Our findings also emphasised the importance of considering how eligibility criteria are operationalised and the impact of complex patient pathways when recruiting to surgical trials. In particular, the influence of health professionals, who are not involved in trial recruitment, on patients’ treatment preferences by suggesting they would receive a certain treatment ahead of recruitment consultations should not be underestimated. Conclusions: A wealth of evidence exploring factors affecting recruitment to randomised controlled trials exists. A methodological shift is now required to ensure that this evidence is used by all those involved in recruitment and to ensure that existing knowledge is translated into methods for optimising recruitment to future trials. Trial registries: ACTIVE: (ISRCTN98152560) PRESTO: (ISRCTN12094890)


2018 ◽  
Vol 29 ◽  
pp. vi14-vi15
Author(s):  
S. Coquerelle ◽  
M. Darlington ◽  
M. Michel ◽  
M. Durand ◽  
J. Gutton ◽  
...  

2021 ◽  
Author(s):  
Stefan O. Kortuem ◽  
Marina Krause ◽  
Hans-Juergen Ott ◽  
Lisa Kortuem ◽  
Hans-Peter Schlaudt

Background. The increasing number of cases and hospital admissions due to COVID-19 created an urgent need for rapid, reliable testing procedures for SARS-CoV-2 in Emergency Departments (ED) in order to effectively manage hospital resources, allocate beds and prevent nosocomial spread of infection. The ID NOW(TM) COVID-19 assay is a simple, user-friendly, rapid molecular test run on an instrument with a small footprint enabling point-of-care diagnostics. Methods. In the first wave, outsourced RT-PCR testing regularly required 36-48 hours before results were available. This prospective study was conducted in the second wave (October 2020-April 2021) and evaluated the impact the implementation of the ID NOW(TM) COVID-19 test in the ED had on clinical care processes and patient pathways. 710 patients were recruited upon arrival at the ED which included those presenting clinical symptoms, asymptomatic individuals or persons fulfilling epidemiological criteria. The first anterior nasal swab was taken by trained nurses in the ambulance or a separate consultation room. The ID NOW(TM) COVID-19 test was performed in the ED in strict compliance with the manufacturer's instructions and positive or suspected cases were additionally tested with RT_PCR (cobas SARS-COV-2 RT-PCR, Roche) following collection of a second nasopharyngeal NP specimen. Results. Swabs directly tested with the ID NOW(TM) COVID-19 test showed a diagnostic concordance of 98 % (sensitivity 99.59 %, specificity 94.55 %, PPV 97.6 %, NPV 99.05 %) compared to RT-PCR as reference. The 488 patients that tested positive with the ID NOW(TM) COVID-19 had a Ct range in RT-PCR results between 7.94 to 37.42 (in 23.2 % > 30). Two false negative results (0.28%) were recorded from patients with Ct values > 30. 14 (1.69%) discordant results were reviewed case-by-case and usually associated with either very early or very advanced stages of infection. Furthermore, patients initially negative with the ID NOW(TM) COVID-19 test and admitted to the hospital were tested again on days 5 and 12: no patient became positive. Discussion. The ID NOW(TM) COVID-19 test for detection of SARS-CoV-2 demonstrated excellent diagnostic agreement with RT-PCR under the above-mentioned patients pathways implemented during the second wave. The main advantage of the system was the provision of reliable results within a few minutes. This not only allowed immediate initiative of appropriate therapy and care for COVID-19 (patient benefit) but provided essential information on isolation and thus available beds. This drastically helped the overall finances of the department and additionally allowed more patients to be admitted including those requiring immediate attention; this was not possible during the first wave since beds were blocked waiting for diagnostic confirmation. Our findings also show that when interpreting the results, the clinical condition and epidemiological history of the patient must be taken into account, as with any test procedure. Overall, the ID NOW(TM) COVID-19 test for SARS-CoV-2 provided a rapid and reliable alternative to laboratory-based RT-PCR in the real clinical setting which became an acceptable part of the daily routine within the ED and demonstrated that early patient management can mitigate the impact of the pandemic on the hospital.


2021 ◽  
Vol 108 (Supplement_9) ◽  
Author(s):  
Martin Michel ◽  
Helen Fifer ◽  
Emily Moran ◽  
Clare Bonner ◽  
Felix Hammett ◽  
...  

Abstract Background The Covid-19 pandemic has affected all aspects of healthcare globally. Theatre utilisation assumes a substantial proportion of hospital resources, creating a streamlined pathway increases efficiency and productivity. With concerns regarding aerosol generating procedures, viral transmission to health care workers in theatre and patient pathways through the hospitals the covid-19 pandemic has added another dimension to the theatre pathway. The aim of this study was to quantify the impact of Covid-19 on the “knife to skin” (KTS) time and compare it to previous historical data (HD).   Methods Retrospective analysis of real time theatre data was analysed for the first 12 months of the pandemic from 11th March 2020 to 11th March 2021. To try and minimise variability between different specialities and operations we picked one operation to study: Laparoscopic cholecystectomy (LC). Historical data was also gathered from the same time frame over the last 5 years (2015-2020) for comparison. Data collected included emergency or elective, time sent for patient, anaesthetic start time, knife to skin time and duration of operation. Comparison of means were analysed by One-way ANOVA tests and Student’s T-Test. Results 399 laparoscopic cholecystectomies were performed during the first year of the pandemic. KTS time was calculated as operation start time minus time sent for patient.  Average time during the pandemic for emergency LC KTS was 56 minutes and 35 minutes for elective LC. Comparison of these times to HD revealed no statistical difference (Emergency LC 56 mins vs 58 mins p > 0.05, Elective LC 35 mins vs 35 mins p > 0.05). The anaesthetic time for emergency LC during the pandemic vs HD was 10 mins vs 14 mins (p < 0.05), no statistical difference was found in the elective group, 16mins vs 14mins (p > 0.05) Conclusions The Covid-19 pandemic has had no detectable effect on Knife to skin time as compared to our previous historical data. It seems the extra Covid 19 precautions involving PPE, pathways etc. have not affected theatre efficiency or utilisation. In fact, there was very little variance in KTS time over the six years studied (2015-2021) with very consistent levels for both elective and emergency procedures. The shorter anaesthetic time for emergency LC during the pandemic needs to be further investigated but one hypothesis is the unconscious or conscious decision to decrease the amount of preoxygenation to minimise aerosolisation.  


2017 ◽  
Vol 131 (5) ◽  
pp. 417-424 ◽  
Author(s):  
L Burrows ◽  
T H Lesser ◽  
A V Kasbekar ◽  
N Roland ◽  
M Billing

AbstractObjective:To report the introduction and impact of non-medical prescribing, initiated to improve patient pathways for those presenting with dizziness and balance disorders.Methods:The Southport and Ormskirk physiotherapy-led vestibular clinic sees and treats all patients with dizziness and balance disorders referred to the ENT department. Letters are triaged by an audiologist, who also performs an otological examination and hearing test; this is followed by an assessment with the independent prescriber physiotherapist. An ENT consultant is nearby if joint consultation is needed. Diagnoses, treatments and patient satisfaction were studied, with an analysis of the impact of medication management (stopping or starting medicines) on patients and service.Results:In 12 months, 413 new patients with dizziness and balance disorders had appointments. The most common diagnoses were benign paroxysmal positional vertigo and vestibular migraine. Eighty-four per cent of patients required self-management strategies, 50 per cent exercise therapy, 48 per cent medication management and 24 per cent a particle repositioning manoeuvre. Patient satisfaction was high (99 per cent).Conclusion:Having an independent prescriber physiotherapist leading the balance clinic has reduced the number of hospital visits and onward referrals. Nearly half of all patients required medication management as part of their dizziness or balance treatment.


Sign in / Sign up

Export Citation Format

Share Document